For More Information

(1)

For More Information

Visit RAND at

www.rand.org

Explore the

RAND Corporation

View

document details

Support RAND

Browse Reports & Bookstore

Make a charitable contribution

Limited Electronic Distribution Rights

This document and trademark(s) contained herein are protected by law as indicated in a notice appearing later in this work. This electronic representation of RAND intellectual property is provided for non-commercial use only. Unauthorized posting of RAND electronic documents to a non-RAND website is prohibited. RAND electronic documents are protected under copyright law. Permission is required from

Skip all front matter:

Jump to Page 1

6 The RAND Corporation is a nonprofit institution that helps improve policy and

decisionmaking through research and analysis.

This electronic document was made available from

www.rand.org

as a public service

of the RAND Corporation.

CHILDREN AND FAMILIES

EDUCATION AND THE ARTS ENERGY AND ENVIRONMENT

HEALTH AND HEALTH CARE INFRASTRUCTURE AND

TRANSPORTATION INTERNATIONAL AFFAIRS LAW AND BUSINESS NATIONAL SECURITY POPULATION AND AGING PUBLIC SAFETY SCIENCE AND TECHNOLOGY TERRORISM AND HOMELAND SECURITY

(2)

This report is part of the RAND Corporation research report series. RAND reports

present research findings and objective analysis that address the challenges facing the

public and private sectors. All RAND reports undergo rigorous peer review to ensure

high standards for research quality and objectivity.

(3)

Christine Eibner

•

Amado Cordova

•

Sarah A. Nowak

• Carter C. Price

Evan Saltzman

•

Dulani Woods

Prepared for the U.S. Department of Health and Human Services

The Affordable Care Act and

Health Insurance Markets

Simulating the Effects of Regulation

(4)

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND’s publications do not necessarily reflect the opinions of its research clients and sponsors.

Support RAND—make a tax-deductible charitable contribution at www.rand.org/giving/ contribute.html

R®

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of RAND documents to a non-RAND website is prohibited. RAND documents are protected under copyright law. Permission is given to duplicate this document for personal use only, as long as it is unaltered and complete. Permission is required from RAND to reproduce, or reuse in another form, any of our research documents for commercial use. For information on reprint and linking permissions, please see the RAND permissions page (http://www.rand.org/pubs/permissions.html).

RAND OFFICES

SANTA MONICA, CA • WASHINGTON, DC

PITTSBURGH, PA • NEW ORLEANS, LA • JACKSON, MS • BOSTON, MA DOHA, QA • CAMBRIDGE, UK • BRUSSELS, BE

The research described in this report was sponsored by the U.S. Department of Health and Human Services. The work was conducted in RAND Health, a division of the RAND Corporation.

(5)

Preface

The Affordable Care Act changes the rating regulations governing the nongroup and small group markets while simultaneously encouraging enrollment through a combination of subsidies, tax credits, and tax penalties. Policymakers and other stakeholders are interested in understanding how these changes might affect health insurance enrollment, premiums, and other outcomes to inform exchange implementation and planning. In this report, we estimate the effects of the Affordable Care Act on health insurance enrollment and premiums for ten states (Florida, Kansas, Louisiana, Minnesota, New Mexico, North Dakota, Ohio, Pennsylvania, South Carolina, and Texas) and for the nation overall, with a focus on outcomes in the nongroup and small group markets. This analysis was sponsored by the Center for Consumer Information and Insurance Oversight (CCIIO, a division of the Centers for Medicare & Medicaid Services [CMS]) through an interagency agreement with the U.S. Department of Health and Human Services (HHS), Assistant Secretary for Planning and Evaluation (ASPE). However, the views, opinions, and findings presented here are those of the authors and should not be construed as official government positions unless so designated by other documents.

The research was conducted by RAND Health, a division of the RAND Corporation. Questions may be addressed to Christine Eibner ([email protected]; (703) 413-‐1100, ext. 5913). A profile of RAND Health, abstracts of its publications, and ordering information can be found at http://www.rand.org/health.

(6)

Summary

The Patient Protection and Affordable Care Act as modified by the Health Care and Education

Reconciliation Act of 2010, collectively known as the Affordable Care Act, makes sweeping changes to the regulation of health insurance markets in the United States.

Specifically, the Affordable Care Act requires insurers in the nongroup and small group markets,

including those offering coverage in the new state-‐level health insurance exchanges, to issue and renew policies to everyone who seeks coverage, regardless of health status. In addition, the Affordable Care Act limits insurers’ ability to charge different prices based on individual characteristics. Insurers can vary prices based only on a few factors:

(1) age

(2) tobacco use

(3) geographic location (4) family size

(5) the actuarial value of the plan.

For these factors, only a certain amount of variation is allowed:

The oldest adult in the risk pool cannot be charged more than three times as much as the youngest adult. This requirement is known as 3-‐to-‐1 rate-‐banding.

In addition, smokers can be charged no more than 1.5 times more than nonsmokers (1.5-‐to-‐1 rate-‐banding).

These changes raise concerns that the Affordable Care Act could lead to substantial increases in premiums, especially in the nongroup market. For example, commentary in the Wall Street Journal published earlier this year suggested that premiums in some markets could double (Matthews and Litow, 2013). Large increases in premiums might occur because of requirements that health insurance be made available to all comers, regardless of health status, and that insurers cannot charge higher premiums based on such characteristics as health status or previous claims experience. Without other changes, these provisions could lead to adverse selection, in which only people with high expected expenditures enroll.

To address these concerns, the Affordable Care Act contains several provisions intended to increase the chances that younger, healthier individuals will get coverage. First, the act requires that all adults obtain a specified minimum level of coverage or pay a tax penalty. Second, the act offers tax credits that individuals with incomes between 100 and 400 percent of the federal poverty level (FPL) can use to buy coverage if they lack access through other sources, such as an employer or Medicaid. Other provisions, such as reinsurance and requirements that insurers limit the amount of premium revenue spent on non-‐ claims costs, could also reduce premiums relative to what would be expected without the law,

(9)

The changes, coupled with other policies introduced by the Affordable Care Act, are likely to affect enrollment, premiums, and the composition of the population enrolled in nongroup and small group plans. It is important for federal and state policymakers to understand the potential effects of these changes as they make decisions about setting up the health insurance exchanges.

In this report, we use RAND’s Comprehensive Assessment of Reform Efforts (COMPARE) microsimulation model to examine these effects. Specifically, our analysis examines the likely effects of the Affordable Care Act on

the number of uninsured individuals

the number of enrollees in the nongroup and small group markets the cost of premiums

the characteristics of enrollees.

We also consider the implications of two decisions confronting states: whether to expand their Medicaid programs to cover all adults with incomes below 138 percent of the FPL and whether to merge or combine their small group and nongroup risk pools. If risk pools are merged, enrollees in the small group and nongroup markets would face the same premiums for comparable coverage. If risk pools are split, premiums in the two markets could diverge. For ten representative states (Florida, Kansas, Louisiana, Minnesota, New Mexico, North Dakota, Ohio, Pennsylvania, South Carolina, and Texas), we estimate enrollment and premiums both with and without the Affordable Care Act. Then, for a subset of states, we conduct sensitivity analyses related to these critical state decisions.

The analysis is based on a microsimulation model, which—like all microsimulation approaches—has limitations. Current data on nongroup premiums are limited, and there are many uncertainties about how individuals and insurers will respond to the complex policy changes introduced by the Affordable Care Act. Nevertheless, state and federal policymakers must continue to implement the law, develop policy guidance and regulations, and make decisions about exchange operations, with little historical data available to gauge the potential effects of these decisions. Recognizing that all models have limitations, our analysis aims to provide decisionmakers with insight into the types of changes in enrollment and premiums that may occur as the Affordable Care Act is implemented.

Results: Number of Uninsured Individuals

Our analysis finds that for all ten states and the United States overall, the Affordable Care Act could

lead to a substantial decline in the number of uninsured nonelderly people. We estimate that the 2016

uninsurance rate in the United States would be 19.6 percent without the Affordable Care Act, compared to 8.2 percent with the law, assuming that all states expand Medicaid. Across states, there is

considerable variation in uninsurance levels in our 2016 estimates, ranging from a low of 5 percent in Minnesota to a high of 12 percent in Texas. States with larger immigrant populations, such as Texas and Florida, tend to have the highest uninsurance rates after Affordable Care Act implementation.

(10)

For three states—Texas, Louisiana, and Florida—we considered the potential consequences for health insurance enrollment in scenarios in which Medicaid was not expanded. Across the three states, we estimate that an additional 2.3 million individuals would be uninsured without Medicaid expansion and state uninsurance rates would increase by 5 to 6 percentage points, compared to scenarios that include Medicaid expansion.

Results: Nongroup Market Enrollment

We estimate that enrollment in the nongroup market will increase substantially across all ten states as a result of the Affordable Care Act. Without the Affordable Care Act, we estimate that fewer than 5

percent of the nonelderly population would be enrolled in nongroup coverage in 2016. With the Affordable Care Act, nongroup enrollment more than doubles, rising from 4.3 percent of the nonelderly population for the United States overall to 9.5 percent of the nonelderly population. The finding that nongroup enrollment could increase, despite the Affordable Care Act’s rating regulations, suggests that adverse selection (the tendency for a disproportionately large number of sicker individuals to opt into the market for health coverage) caused by these new regulations is mitigated by other provisions. For example, the individual mandate and federal tax credits for exchange enrollees with incomes between 100 and 400 percent of the FPL could keep younger and healthier people enrolled.

Results: Nongroup Premiums

The results for premium prices in the nongroup market are complicated and must be interpreted carefully because the law introduces complex changes and because of limitations of existing data and

uncertainties about insurer behavior. The law’s requirement that individuals obtain plans with a minimum actuarial value will cause some enrollees to shift from less-‐generous into more-‐generous plans, which could result in higher premiums but also more-‐comprehensive coverage. In addition, some individuals could experience declines in out-‐of-‐pocket premiums even if their total premiums increase, due to eligibility for federal premium tax credits. Because the Affordable Care Act allows insurers to charge higher premiums to older individuals (within a 3-‐to-‐1 rate band) and tobacco users (within a 1.5-‐ to-‐1 rate band), the change in average premiums could be different from the change in premium for an individual with a fixed age and tobacco use status. Finally, data currently available on nongroup

premiums and enrollment are limited and vary substantially across sources, which affects the reliability of all predictions, including estimates presented in this report.

In analyses that held age, actuarial value, and tobacco use constant, we estimated that, for five of the ten states we examined (Florida, Kansas, Pennsylvania, South Carolina, and Texas), and for the United States overall, the law causes no change in premiums.1_{In
three
of
the
remaining
states
(Minnesota,}

North Dakota, and Ohio), we estimated that there could be premium increases of up to 43 percent, although these changes do not account for federal exchange tax credits. For two states (Louisiana and New Mexico), we estimated that premiums standardized for age, actuarial value, and tobacco use could

1_{While
premium
estimates
differed
with
and
without
the
law,
we
could
not
reject
the
possibility
that
these}

(11)

decline as a result of the Affordable Care Act. Declines in premiums are possible in part because premium tax credits bring some people who are relatively less expensive into the market, reinsurance reduces nongroup premiums in the first few years of implementation, and the Affordable Care Act limits the amount of premium revenue that can be spent on non-‐claims costs. Even among states estimated to have an increase in total premiums standardized for age, actuarial value, and tobacco use, many

enrollees will experience a decline in their out-‐of-‐pocket premium expenditures because part of the premium is subsidized via federal tax credits.

For three states (Texas, Louisiana, and Florida), we considered the effect of Medicaid expansion on nongroup premiums. If states fail to expand Medicaid, individuals with incomes in the range of 100 to 138 percent of the FPL will become newly eligible for exchange tax credits. We find that, for these three states, these newly eligible individuals could cause premiums standardized for age, actuarial value, and tobacco use on the nongroup market to rise by 8 to 10 percent, relative to scenarios that include Medicaid expansion. The increase in premiums reflects an influx of slightly lower-‐income and less-‐ healthy enrollees onto the exchanges.

Results: Small Group Market Enrollment

For the United States overall, and for seven of ten states (Florida, Louisiana, Minnesota, New Mexico, Ohio, South Carolina, and Texas), we estimate that small group enrollment will be larger in scenarios that include the Affordable Care Act, with increases ranging from less than 1 to approximately 5 percentage points. Three states—Kansas, North Dakota, and Pennsylvania—are estimated to experience

modest declines in small group coverage, ranging from 1.4 to 2.2 percentage points.

The finding that states could experience an increase in small group enrollment reflects the fact that workers will have increased demand for health insurance as a result of the act, due to penalties

associated with not having insurance. In addition, although many lower-‐income workers will be eligible for exchange tax credits or Medicaid, higher-‐income workers benefit from the tax advantage associated with employer-‐sponsored coverage. Firms must make a single health insurance offering decision for all workers, and—for many businesses—the tax benefits to higher-‐income workers could dominate decisionmaking.

However, a limitation in our analytic framework is that it does not allow us to consider whether firms reduce workers’ hours, change premium contribution rates, or alter their sizes in response to the law. This type of strategic response could lead small group enrollment under the Affordable Care Act to be lower than estimated if, for example, firms convert some workers to part-‐time status to avoid offering coverage.

Results: Small Group Premiums

We find minimal difference in small group premiums in scenarios with and without the Affordable Care Act. For the United States overall, and for nine of ten states (Florida, Kansas, Louisiana, Minnesota,

(12)

standardized for age, actuarial value, and tobacco use will be unchanged by the law.2_{These
scenarios}

assume that the nongroup and small group risk pools are separated.

For one state that was not included in our group of ten (New York), we analyzed the potential effects of merging the small group and individual market risk pools. In this case, we found that small group premiums standardized for age, actuarial value, and tobacco use could be as much as 16 percent higher if the small group and nongroup markets are merged, relative to the case in which the markets are separated. Both the separated and combined risk pools scenarios assume that the Affordable Care Act is fully in effect. New York may be an unusual case because it has unusually strong regulations in both its nongroup and its small group markets.3_{However,
we
estimate
that—for
all
states
considered
in
our}

analysis—nongroup enrollees in scenarios with the Affordable Care Act will be slightly older and less healthy than small group enrollees. These estimates imply that, for many states, small group premiums could increase if the nongroup and small group risk pools are combined.

Conclusions

We conclude that the Affordable Care Act will lead to an increase in insurance coverage and higher enrollment in the nongroup market. However, data limitations and uncertainties about insurer

behavior make estimates uncertain, particularly when considering outcomes for the nongroup market. We find that the law has little effect on small group premiums, and we find large variation in the effects for nongroup premiums across states.

Our analysis suggests that comparisons of average premiums with and without the Affordable Care Act may overstate the potential for premium increases. Sweeping statements about the effects of the

Affordable Care Act on premiums should be interpreted very carefully because the law has complex effects that will differ depending on individuals’ age and smoking status, the actuarial value of the plan chosen, individuals’ eligibility for federal tax credits, and state implementation decisions. Once we adjust for age, actuarial value, and tobacco use, nongroup premiums are estimated to remain unchanged at the national level and in many states. Further, after accounting for tax credits, average out-‐of-‐pocket premium spending in the nongroup market is estimated to decline or remain unchanged in all states considered and in the nation overall.

2_{That
is,
statistically,
we
cannot
reject
the
hypothesis
that
premiums
are
equivalent
in
scenarios
with
and
without}

the law.

3_{New
York
has
full
community
rating
in
both
its
small
group
and
nongroup
markets,
meaning
that
all
enrollees}

(13)

Acknowledgments

We are grateful for guidance and comments provided by stakeholders at ASPE and CCIIO, including John Bertko, Nancy DeLew, Ken Finegold, Sherry Glied, Beth Hadley, Rick Kronick, Ben Lum, Dena Puskin, and Laura Skopec. Their feedback and support greatly enhanced the analysis presented here.

We are additionally indebted to three technical reviewers, Jean Marie Abraham of the University of Minnesota, Matthew Buettgens of the Urban Institute, and Raffaele Vardavas of RAND, who provided insightful comments on an early draft of the report. We also thank Robin Weinick, Rosalie Pacula, and Paul Koegel for their comments on draft versions of the document.

Finally, we thank Stacy Fitzsimmons for providing excellent administrative assistance in preparing this report and Alan Weaver for his expert help in navigating contractual aspects of the project.

(14)

I. Introduction

The Patient Protection and Affordable Care Act as modified by the Health Care and Education

Reconciliation Act of 2010, collectively known as the Affordable Care Act, was signed into law on March 23, 2010, and represents the biggest change to the U.S. health care system since Medicare was enacted in 1965. The act expands the Medicaid program, requires most individuals to obtain insurance or pay a penalty, provides subsidies to individuals who have low to moderate incomes and no affordable source of coverage, and imposes fines on business with more than 50 employees that do not offer adequate coverage to their workers if those workers seek federally subsidized coverage as an alternative. To ensure that all individuals and small businesses are able to obtain health insurance policies, the Affordable Care Act also introduces new federal regulations in the nongroup (privately purchased) and small group (small employer) health insurance markets that limit insurers’ ability to deny coverage or to charge differential prices based on certain enrollee characteristics. Although there are some exceptions for plans that existed prior to the Affordable Care Act’s enactment and have not made substantial changes to cost-‐sharing or other requirements over time, most small group and nongroup plans will be required to adhere to the new regulations.

Separately, to facilitate competitive shopping for qualified insurance plans, the law encourages states to set up and run health insurance exchanges for the small group and nongroup insurance markets. These exchanges—designated the Health Insurance Marketplaces (HIMs)4_{for
individuals
and
the
Small
Group}

Health Insurance Options Program (SHOP) Exchanges for firms—will include online portals to help individuals and small employers shop for, select, and enroll in health insurance plans and will offer subsidies and tax credits to qualified individuals and firms. While states are encouraged to implement health insurance exchanges on their own, they may allow this administrative role to fall to the federal government if they decline to establish a Marketplace, and some states are setting up state and federal partnership exchanges. Whether the exchanges are state-‐run or federally run, those tasked with setting up exchanges must make many regulatory and implementation decisions. Two significant decisions that states are currently grappling with include whether to combine the small group and nongroup markets for the purposes of risk pooling and whether to expand their Medicaid programs to cover all adults with incomes below 138 percent of the federal poverty level (FPL).

While the Affordable Care Act initially required states to expand Medicaid to all individuals with incomes below 138 percent of the FPL, the Supreme Court ruled in June 2012 that a requirement to expand was unduly coercive, and states may therefore opt out of the expansion of Medicaid if they prefer. The decision to opt out of the Medicaid expansion will have implications for exchange enrollment, since individuals with incomes between 100 and 138 percent of the FPL will become eligible for federal exchange subsidies—formally known as advance premium tax credits (APTCs)—if states opt out of the Medicaid expansion. APTCs are not generally available for individuals with incomes below 100 percent

(15)

of the FPL, although there are exceptions for recent immigrants who would not qualify for Medicaid even if the state expanded.

Because the federal government must establish regulations and provide guidance to states that choose to operate their own exchanges, and because some states may opt to default to the federally facilitated exchanges, policymakers at both the state and federal levels may need estimates of potential exchange-‐ related outcomes to facilitate planning. Key outcomes of interest include the number of people

projected to participate in exchanges, the number of enrollees who are subsidy-‐eligible, exchange premium prices, and aggregate federal spending amounts. In this report, we use RAND’s Comprehensive Assessment of Reform Efforts (COMPARE) microsimulation model to analyze these outcomes, focusing on ten representative states—Florida, Kansas, Louisiana, Minnesota, New Mexico, North Dakota, Ohio, Pennsylvania, South Carolina, and Texas. These states were selected by our client, the Centers for Medicare & Medicaid Services’ (CMS’s) Center for Consumer Information and Insurance Oversight (CCIIO). For sensitivity analyses, we added an eleventh state: New York.

Core questions addressed in this report include the following:

How many individuals and employees of small businesses are likely to obtain coverage on the newly regulated small group and nongroup markets? Of those participating, how many are eligible for exchange-‐based subsidies or tax credits?

What are the risk profiles, income levels, and prior coverage patterns of potential exchange enrollees?

What are the projected premiums for exchange plans, including both the HIMs and the SHOP exchanges?

We also conduct sensitivity analysis to explore the potential consequences of states opting out of the Medicaid expansion for exchange enrollment and premiums and to address the implications of decisions regarding risk-‐pool composition on nongroup and SHOP premium prices.

(16)

II. The Affordable Care Act and the Nongroup and Small Group

Markets

The Affordable Care Act changed the laws governing how health insurers set premiums in the nongroup and small group markets. The basic function of insurance is to pool risk across a group of individuals: Each enrollee contributes money to the pool in the form of a premium, and this money is then redistributed to pay for enrollees’ medical care. If the chance of having a high expenditure were unpredictable and did not vary systematically with such characteristics as age and health status, then everyone would be charged the same premium. However, because health spending is somewhat predictable, insurers have an incentive to charge higher premiums (or deny coverage entirely) to

individuals who are older, sicker, or otherwise expected to be high spenders. A typical argument in favor of this type of differential pricing is that younger and healthier people will not agree to pay the

premiums required to subsidize spending on older and sicker individuals. As a result, if premiums are not differentiated based on age and health status, only people with high anticipated spending will remain in the pool, an outcome known as “adverse selection.”

Large variation in premiums can be problematic from a policy perspective if the goal is to ensure that all individuals have access to affordable coverage. As a result, regulatory interventions have sometimes been used to limit insurers’ ability to deny coverage or to charge substantially higher premiums to older and sicker individuals. Prior to the Affordable Care Act, the Health Insurance Portability and

Accountability Act (HIPAA) prohibited insurers from denying coverage to enrollees in small group plans (for employers with 50 or fewer workers), but such regulations often did not exist in the nongroup market, which was governed primarily by state law. Most states also allowed insurers to charge nongroup enrollees different premiums depending on age, health status, and gender. The handful of states that restricted insurers’ ability to charge differential prices in the nongroup market, including New York and New Jersey, tended to experience a reduction in coverage among younger or healthier

enrollees, a hallmark of adverse selection (Pauly and Herring, 2007; Lo Sasso and Lurie, 2009).

The Affordable Care Act introduced new federal requirements in both the small group and the nongroup markets that require insurers to issue and renew policies to all comers regardless of health status. In addition, the act imposed strict limitations on insurers’ ability to price-‐discriminate. Specifically, the Affordable Care Act allows issuers to charge different prices based only on a small subset of factors: age, tobacco use, geographic location, family size, and plan actuarial value. Within these factors, there are limits on the amount of price variation that is permissible. Although insurance companies can rate on age (that is, they can vary premiums based on enrollee age), the oldest adult in the risk pool cannot be charged more than three times as much as the youngest adult, a policy known as 3-‐to-‐1 rate-‐banding. Similarly, smokers can be charged no more than 1.5 times as much as nonsmokers (1.5-‐to-‐1 rate-‐ banding).

(17)

Rating regulations, such as those required under the Affordable Care Act tend to increase premiums for younger and healthier enrollees while lowering premiums for older and sicker enrollees. As a result, the regulations raise the concern that only the sickest individuals will opt to purchase health insurance. However, the Affordable Care Act contains several provisions that are likely to increase the chances that young and healthy people enroll and remain enrolled:

1. The law mandates than all individuals obtain insurance and imposes tax penalties on most individuals who do not have qualifying coverage.

2. The Affordable Care Act offers tax credits to individuals who have incomes between 100 and 400 percent of the FPL and who do not have access to affordable coverage through other sources. 3. The Affordable Care Act sets limits on the medical loss ratio (MLR), limiting the share of

premium payments that health insurers can use to finance non–claims-‐related administrative costs and certain other specified activities, such as quality improvement and effort to detect fraud and abuse. To the extent that the MLR restrictions reduce wasteful administrative spending, they may make premiums less expensive for all enrollees.

Tax credits for individuals are available only if they enroll in coverage through the newly created nongroup health insurance exchanges. The exchanges are new marketplaces for buying and selling insurance in the nongroup and small group markets. In addition to the individual tax credits, the act establishes temporary tax credits for businesses with 25 or fewer workers who have average wages below $50,000 per year; these tax credits can only be applied to exchange-‐based coverage. The Affordable Care Act envisioned that the exchanges would be set up and run by states but also allowed for federally facilitated or state and federal partnership exchange options. Although pricing regulations are the same both within and outside of the exchanges, exchange plans will be subject to at least ten additional provisions (Jost, 2010), including requirements that plans include providers that serve low-‐ income, medically needy populations; that plans be accredited on the basis of the Healthcare

Effectiveness Data and Information Set (HEDIS) and the Consumer Assessment of Healthcare Providers and Systems (CAHPS); and that plans use a uniform enrollment form.

The Affordable Care Act also contains several provisions that are designed to limit adverse selection across plans—a process in which insurers would steer high-‐cost enrollees into particular plans to effectively segment the market based on health status. First, the entire small group market—regardless of plan—must be considered a single risk pool by a health insurance issuer in a given state. Similarly, the entire nongroup market must be considered a single risk pool. Second, risk adjustment is required to transfer funds from plans with “higher than average” actuarial risk to those with “lower than average” risk. Together, these provisions are intended to help stabilize premiums and keep more-‐generous plans affordable, despite the fact that more-‐expensive individuals may gravitate toward these plans.

The law also includes temporary provisions that are intended to stabilize the nongroup market in the early years of the Affordable Care Act (2014 through 2016). All insurers will be required to make payments to a reinsurance fund, which will then be used to compensate plans in the nongroup market for the costs of enrollees with high realized expenditures. This provision helps to reduce premiums in the nongroup market. The law also establishes risk corridors for small and nongroup plans in the initial

(18)

years that limit profits and losses. The risk corridors protect insurers against the possibility that they are off target in setting premiums—an outcome that could be likely, given a lack of historical experience with the newly insured population.5_{With
the
exception
of
reinsurance
payments,
most
of
the
provisions}

described above apply to all nongrandfathered small group and nongroup plans.

(19)

III. COMPARE Background

COMPARE is a microsimulation model that estimates the effects of health policy changes at the national and state levels on health insurance enrollment decisions and premiums. The model uses a synthetic dataset with information on a nationally representative sample of individuals and their employers and estimates how they will react under different policy scenarios. In COMPARE, we used data from the Survey of Income and Program Participation (SIPP), the Medical Expenditure Panel Survey (MEPS), and the Kaiser Family Foundation/Health Research and Educational Trust (Kaiser/HRET) Survey of Employer Benefits to create a synthetic population of individuals, families, and firms that represents the overall U.S. population. These individuals, families, and firms then make simulated health insurance choices by weighing the costs and benefits of available options. We calibrate the model so that it accurately reproduces health insurance enrollment decisions under pre–Affordable Care Act policy. We then introduce the policy changes enacted under the Affordable Care Act. The major policy changes simulated in our model include

the expansion of Medicaid to all individuals with incomes below 138 percent of the FPL (allowing for states to opt out of the expansion in sensitivity testing)

the creation of health insurance exchanges in the individual and small group markets

federal APTCs for individuals with incomes between 100 and 400 percent of the FPL who do not have affordable coverage from another source, such as Medicaid or employer coverage

cost-‐sharing subsidies for people with incomes between 100 and 250 percent of the FPLwho do not have affordable coverage from another source

new rating regulations in the small and nongroup insurance markets

penalties for employers and individuals who do not offer or enroll in coverage.

As described below, we also model many of the laws’ more-‐detailed provisions, such as reinsurance for nongroup plans, and new federal requirements regarding medical loss ratios.

Individual Choices

COMPARE uses a utility-‐maximization approach to simulate the decisionmaking of individuals and families—or, more specifically, the health insurance eligibility units (HIEUs)—with respect to purchasing health insurance. On the assumption that individuals will change their insurance status based on

deductive rational thinking, an HIEU selects the set of insurance policies that maximize the utility of each member. We assume that the utility function is quasilinear and separable in health and other

consumption. The health-‐related component of the utility function is defined as

(20)

where u(Hij) is the utility associated with consuming health care services for individual i under insurance

option j,6_OOP_{ij
is
the
out-‐of-‐pocket
spending
expected,
p}(H)_{is
the
premium,
and
r
is
the
coefficient
of}

risk aversion. The insurance status j will depend on the options available to each individual but includes employer-‐sponsored coverage (for those with an employer offer), nongroup coverage, Medicaid (for those meeting eligibility criteria), or uninsurance. In the post–Affordable Care Act environment, insurance options can also include bronze, silver, gold, or platinum plans offered on the exchanges. HIEUs weigh the benefits of an option (e.g., reduced out-‐of-‐pocket expenditure, lower risk) against the costs (e.g., higher premiums).

To estimate out-‐of-‐pocket spending, we first design synthetic plans, defined by a deductible,

coinsurance, and maximum out-‐of-‐pocket spending, for each plan type. In the status quo, we have plans for the uninsured, Medicaid, employer-‐sponsored insurance, and the nongroup market. The uninsured “plan” has no premium but accounts for the out-‐of-‐pocket spending that individuals would be likely to face if not insured. Under the Affordable Care Act, we have plans for the uninsured, Medicaid, large group employer-‐sponsored insurance, each metal tier in the regulated small group and nongroup markets (bronze, silver, gold, and platinum plans), and each level of affordable cost-‐sharing subsidies available on the nongroup market. We designed synthetic plans for each insurance type to achieve the specified actuarial value of each plan. Status quo plans were additionally chosen to reflect out-‐of-‐pocket and total health care spending reported in the MEPS. We then used these synthetic plans to estimate each individual's out-‐of-‐pocket spending from estimated total health care spending.

In making health insurance decisions, we assume that individuals and families will choose the option that maximizes their health-‐related utility. HIEUs consider an array of factors, including eligibility for Medicaid, eligibility for subsidies on the health insurance exchange, the generosity of the plan they are considering, health insurance premiums, tax penalties for not obtaining coverage, and expected health expenditures. Subject to a few constraints (e.g., a child cannot be enrolled in employer-‐sponsored insurance if a parent is not also enrolled), the HIEU can select different options for each member. We do not account for the possibility that there could be coordination costs associated with having family members enrolled in different plans—for example, costs caused by added paperwork requirements. When the Affordable Care Act takes effect, penalties associated with not having health insurance coverage enter into the utility function. Specifically, these penalties are subtracted from the utility associated with the option of being uninsured.

Firm Choices

Firms in COMPARE maximize the aggregate utility of their workers, enabling them to make the health insurance decision that provides the best value to the most workers. In some cases, the optimal decision could be to not offer insurance or to drop health insurance coverage. Following standard economic

6_{We
assume
that
the
individual
experiences
an
increase
in
utility
from
consuming
additional
health
care
services,}

potentially because additional health care consumption improves health. However, the specification is agnostic as to why individuals prefer higher health care consumption and would still be accurate if individuals get utility from consuming health care services that do not improve their ultimate health outcomes.

(21)

theory (Baicker and Chandra, 2008), we assume that workers face a trade-‐off between health insurance and wages, so that wages fall as health insurance costs increase, and vice versa. The wage–health insurance trade-‐off assumption implies that, if the firm opts to stop offering health insurance, wages will have to increase. In a purely competitive market, wages would adjust one-‐for-‐one with changes in employer health insurance expenditure, and several empirical studies have found that workers bear the full incidence of higher benefits costs through reduced wages (Bhattacharya and Bundorf, 2009; Gruber, 1994). However, prior research has also suggested that nominal wages could be sticky in the short run (Sommers, 2005), causing wages to fall by less than the change in employer health expenditure. In the COMPARE model, we assume that for every dollar reduction in health care spending, the firm will pass 80 cents back to workers. However, in prior work, we have shown that the model results are insensitive to wage pass-‐back assumptions, particularly for wage pass-‐back rates of 70 percent or higher (Eibner et al., 2010).

In determining whether they would prefer an offer of health insurance to a change in wages, firms consider the value of alternative health insurance options available to their workers, which depends on whether they are eligible for Medicaid or exchange tax credits in the absence of a firm offer, the tax advantage associated with employer-‐sponsored coverage, the utility of health insurance described above, and the penalty that will be imposed if workers do not enroll in coverage.

In implementing the firm decisionmaking process, we make several simplifying assumptions: 1. We assume that firms offer, at most, one plan. According to the most recent data from

Kaiser/HRET (2012), 82 percent of all offering firms offer only one plan, and small firms—the focus of this report—are more likely to offer a single plan than larger firms.

2. We estimate employee premium contribution rates using a regression, which is based on the Kaiser/HRET data.

We do not allow for the possibility that firms might reduce worker hours or change size in response to the Affordable Care Act. A more detailed description of the firm behavior algorithm can be found in two previous reports (Eibner et al., 2010; Eibner et al., 2011).

Premiums

Premiums in the COMPARE model are calculated based on the expenditure of enrollees in the pool, with adjustments for plan actuarial value and administrative costs. In scenarios without the Affordable Care Act we allow premiums to vary based on enrollee health status in both the small group and the nongroup markets, unless prohibited by state law. In the nongroup market, we also account for the possibility that some individuals will be denied coverage because of high expected expenditures. After 2014, we allow premiums in the nongroup and small group markets to vary only by age (with 3-‐to-‐1 rate-‐banding for adults), tobacco use status (with 1.5-‐to-‐1 rate-‐banding), family size, geography, and plan actuarial value. We also require guaranteed issue and guaranteed renewal of plans—that is, insurers may not deny coverage to potential enrollees.

For More Information